A touch interface for soft data modeling in Bayesian estimation

Mehta, Siddhartha S.; McCourt, Michael; Doucette, Emily A.; Curtis, J. Willard

doi:10.1109/smc.2014.6974511

Cited by 9 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kaupp et al [6] and Wang et al [10] extend the core concept one step further to consider how uncertainties in human sensor model parameters can be accounted for in data fusion using maximum likelihood estimation, whereas Ahmed et al [4] considers a fully Bayesian reasoning approach. Dani et al [11], Mehta et al [12], and Bishop et al [13] have also considered alternative human-machine communication interfaces and probabilistic models for soft data fusion in dynamic target tracking problems. However, a key assumption for data modeling and fusion in all these works is that observations from a single observation remain i.i.d.…”

Section: Background and Related Workmentioning

confidence: 99%

Fully Bayesian Human-Machine Data Fusion for Robust Dynamic Target Surveillance and Characterization

Muesing

Burks

Iuzzolino

et al. 2019

AIAA Scitech 2019 Forum

View full text Add to dashboard Cite

Section: Background and Related Workmentioning

confidence: 99%

Fully Bayesian Human-Machine Data Fusion for Robust Dynamic Target Surveillance and Characterization

Muesing

Burks

Iuzzolino

et al. 2019

AIAA Scitech 2019 Forum

View full text Add to dashboard Cite

“…7,8 While humans are not able to observe and report measurements as quickly as automated sensors, they outperform algorithms in terms of situational awareness. This includes recognizing threats in an environment.…”

Section: [ Imentioning

confidence: 99%

“…8 An alternative approach to using humans as sensors is to allow each human to report measurements graphically. 7 This can be done using a touch-based interface where the person directly manipulates a point cloud that represents an assessment of the target state. This requires that the human has some basic level of training so that the person is able to accurately represent his or her uncertainty in the measurement by appropriately drawing the point cloud.…”

Section: Human Assisted Estimationmentioning

confidence: 99%

Human-machine teaming for effective estimation and path planning

et al. 2016

View full text Add to dashboard Cite

While traditional sensors provide accurate measurements of quantifiable information, humans provide better qualitative information and holistic assessments. Sensor fusion approaches that team humans and machines can take advantage of the benefits provided by each while mitigating the shortcomings. These two sensor sources can be fused together using Bayesian fusion, which assumes that there is a method of generating a probabilistic representation of the sensor measurement. This general framework of fusing estimates can also be applied to joint human-machine decision making. In the simple case, binary decisions can be fused by using a probability of taking an action versus inaction from each decision-making source. These are fused together to arrive at a final probability of taking an action, which would be taken if above a specified threshold. In the case of path planning, rather than binary decisions being fused, complex decisions can be fused by allowing the human and machine to interact with each other. For example, the human can draw a suggested path while the machine planning algorithm can refine it to avoid obstacles and remain dynamically feasible. Similarly, the human can revise a suggested path to achieve secondary goals not encoded in the algorithm such as avoiding dangerous areas in the environment.

show abstract

“…This avoids incorrectly biasing or collapsing state probability distributions in the presence of imperfect human sensing, thereby improving collaborative human-robot sensing for object/world state estimation. PSDA naturally integrates with existing hybrid 'soft-hard' Bayesian data fusion schemes developed in earlier work for collaborative human-robot team sensing [5], [6], [10]- [13]. Furthermore, although PSDA requires knowledge of additional human sensing characteristics like false positive rates, it is robust to modest mismatches between assumed and actual human sensor parameters.…”

Section: Introductionmentioning

confidence: 99%

Probabilistic Semantic Data Association for Collaborative Human-Robot Sensing

Wakayama¹,

Ahmed

2021

Preprint

View full text Add to dashboard Cite

In collaborative human-robot semantic sensing problems, e.g. for scientific exploration, robots could potentially overtrust information given by a human partner, resulting in suboptimal state estimation and poor team performance. When humans cannot be treated as oracles, robots need to update state beliefs to correctly account for possible discrepancies between human semantic observations and the actual world states which lead to those observations. This work develops strategies for rigorous online calculation of probabilistic semantic data association (PSDA) probabilities for semantic likelihoods in general settings, unlike previous work which developed naive or heuristic approximations for specific settings. The new PSDA method is incorporated into a hybrid Bayesian data fusion scheme which uses Gaussian mixture priors for object states and softmax functions for semantic human sensor observation likelihoods, and is demonstrated in Monte Carlo simulations of collaborative multi-object search missions featuring a range of relevant human sensing characteristics (e.g. false detection rate). It is shown that PSDA leads to robust estimation of observation association probabilities under a wide range of conditions whenever semantic human sensor data contain significant target reference ambiguities for autonomous object search and localization.

show abstract

A touch interface for soft data modeling in Bayesian estimation

Cited by 9 publications

References 17 publications

Fully Bayesian Human-Machine Data Fusion for Robust Dynamic Target Surveillance and Characterization

Fully Bayesian Human-Machine Data Fusion for Robust Dynamic Target Surveillance and Characterization

Human-machine teaming for effective estimation and path planning

Probabilistic Semantic Data Association for Collaborative Human-Robot Sensing

Contact Info

Product

Resources

About